Determination of the N-Nitrosodimethylamine Content, pp.5-6 Volume 106, No. 1, 2000

Determination of the N-Nitrosodimethylamine

Content of Beer and Malt - A Collaborative Trial

Submitted by V. Kellner on behalf of the Analysis Committee of the European Brewery Convention

After about fifteen years a study was carried out to determine the variation in results of the

N-nitrosodintethylamine (NDMA) determination in beer and malt carried out in different

laboratories, in order to obtain an overall view of the precision of the various methods used.

Two collaborative trials were carried out by the EBC Analysis Committee during 1998.

Repeatability values (rg5)for beer in the range of 0.09 + 0.13 andfor malt 0.33 + 0.42 were found.

Reproducibility values (R95)for beer in the range of 0.45 + 0.54 and for malt 0.65 -*- 4.08 were

obtained.

Key Words: Collaborative trial, ring test, NDMA, N-nitrosodimethylamine (determination of), beer, malt.

INTRODUCTION

In April 1998 the first collaborative trial was carried out

using three beer samples (two spiked with NDMA) and

one commercial malt. Six participants took part in the

trial on beer but only five returned malt results.

Furthermore, after the statistical treatment the number

of acceptable results was yet lower and so it was decided

to repeat the collaborative trial with more participants.

The second trial was carried out in October 1998.

EXPERIMENTAL

The organization of the collaborative trials and the

statistical treatment of the data were carried out

according to the procedures given in International

Standards ISO 5725-1 and ISO 5725-21-2. A uniformly

designed trial was employed and six samples were

circulated to 11 participants: three beers (all spiked with

NDMA; original beer 0.3 ug/1 NDMA) and three

commercial malts without NDMA additions. Laboratories

were asked to determine the NDMA concentrations in

duplicate to 1 decimal place. Beer results were obtained

from six participants, malt results were submitted by 9

laboratories. Two laboratories did not send their results;

one of them had been excused due to the TEA detector

being out of order.

RESULTS AND DISCUSSION

The precision data from the first collaborative test

are given in Table I for illustration only. Original data

of the second collaborative trial are given in Table II.

The participants used MEBAKmethods45, ASBC

methods6, Drost method3, or their own methods/

modifications. For beer 1, beer 2 and malt A no

outliers or stragglers were identified using Cochran's

and Grubbs' test. Laboratory I for beer 3 was

identified as a straggler using Grubbs' test. The same

laboratory was identified as a straggler for malt B and

an outlier for malt C using Cochran's test. The

precision data are summarized in Table III. Values of

r95 for the beer are in the range of 0.09 •+• 0.13 and for

malt 0.33 + 0.42. Values of R95 for beer are in the range

of 0.45 + 0.54 and for malt 0.65 + 4.08. Comparing

these results with those obtained in 1982/ 19833 We

see that they are nearly identical. The situation in the

field of the N-nitrosodimethylamine determination

remains unchanged.

TABLE I. Summary of the precision data of the first trial (April 1998) [ppb] CONCLUSIONS

Level

Sample

Number of laboratories

Grand mean

Repeatability standard

deviation

Reproducibility

standard deviation

Repeatability r95

Reproducibility R^

1

Beerl

5

0.11

0.003

0.013

0.009

0.036

2

Beer 2

5

0.65

0.021

0.052

0.059

0.146

3

Beer 3

4

1.16

0.036

0.111

0.101

0.311

4

Malt

4

0.97

0.035

0.085

0.098

0.238

ppb... [ng/kgor(ig/1]

Beer 1: no addition of NDMA; Beer 2: Beer 1 + 0.5 Jlg/l NDMA;

Beer 3: Beer 1 + 1.0 ng/1 NDMA

The Analysis Committee of the

European Brewery Convention carried

out two collaborative trials for the

determination of the NDMA content in

beers and malts. Participants used

MEBAK methods, ASBC methods, the

Drost method or in-house methods. The

results of the trials indicate that the

precision and accuracy of the

methodology used is satisfactory.

Therefore, there is no requirement to

establish a harmonized methodology.

Journal of The Institute of Brewing

This document is provided compliments of the Institute of Brewing and Distilling www.ibd.org.uk Copyright - Journal of the Institute of Brewing

Volume 106, No. 1,2000 Determination of the N-Nitrosodimethylamine Content

TABLE II. Original data of the second collaborative trial (October 1998)

Sample

Lab

A

B

C

D

E

F

G

H

I

Mean°

GrandMean

Beer 1

1.2

-

-

1.2

1.0

0.98

-

12

0.77

1.06

1.1

-

-

1.2

1.0

0.95

-

1.1

0.70

1.01

1.03

Beer 2

0.7

-

-

0.6

0.6

0.35

-

0.7

0.32

0.55

0.7

-

-

0.6

0.6

0.31

-

0.6

0.36

0.53

054

Beer 3

1.1

-

-

1.0

1.1

0.92

-

1.0

0.58*

0.95

1.1

-

-

1.0

1.1

0.94

-

1.1

0.61*

0.97

0.96

Malt A

1.1

1.4

0.7

0.9

1.0

1.2

0.9

0.8

1.27

1.03

1.05

1.4

0.7

0.9

1.1

1.1

1.1

0.7

0.84

0.99

1.01

MaltB

2.2

3.3

1.3

2.2

2.1

23

23

2.0

2.82+

2.28

2.4

33

1.3

2.1

2.1

2.5

2.4

2.0

2.2T+

2.26

2.27

MaltC

4.6

7.5

2.2

5.1

4.5

4.5

5.1

4.5

435**

4.69

4.7

75

2.2

55

4.5

4.4

5.3

4.4

5.16++

4.85

4.77

ppb... [ug/kgorug/1]

Straggler at p = 0.05 based

on Cochran's test

++ Outlier at p = 0.01 basedon Cochran's test

° Mean excluding the outlier(s)

Spiking:

Original beer X = 03 ug/1 ppb

Beer 1: X + 0.85 ug/1 NDMA; Beer 2; X + 0.35 ug/I NDMA;

Bcer3:X + 0.75ug/l

• Straggler at p = 0.05

based on Grubbs' test

" Outlier at p = 0.01

based on Grubbs' test

4. Brautechnische Analysenmethoden,

Band in, S. 435-443. MEBAK, Freising-

Weihenstephan 1982.

5. Brautechnische Analysenmethoden,

Band III, S. 50-53.2. Auflage MEBAK,

Freising-Weihenstephan 1996.

6. Methods of Analysis of the American

Society of Brewing Chemists, Malt-10,

Beer-40. 8th Edition. ASBC, Inc., St.

Paul 1992.

TABLE III. Summary of the precision data of the second trial (October 1998)

Ippb]

Level

Sample

Number of

laboratories

Grand mean

Repeatability

standard deviation

Reproducibilitystandard deviation

Repeatability r95

Reproducibility R95

1

Beerl

6

1.03

0.046

0.18

0.13

0.49

2

Beer 2

6

0.54

0.033

0.16

0.09

0.45

3

Beer 3

6

0.96

0.031

0.19

0.09

0.54

4

Malt A

9

1.00

0.012

0.23

0.34

0.65

5

MaltB

9

2.27

0.15

0.54

0.42

1.50

6

MaltC

8

4.77

0.12

1.46

0.33

4.08

ppb... Ing/kg or ug/1]

REFERENCES

1. International Standard, Accuracy (trueness and

precision) of measurement methods and results -

Part 1: General principles and definitions, ISO 5725-

1,1994.

2. International Standard, Accuracy (trueness and

precision) of measurement methods and results -

Part 2: Basic method for the determination of

repeatability and reproducibility of a standard

measurement method, ISO 5725-2,1994.

3. Pfenninger, H. and Anderegg, P., /. Inst. Brew., 1984,90

(6), 338; Pfenninger, H. and Anderegg, P., Monatsschr.

Brauwiss, 1984, 37 (11), 478; Pfenninger, H. and

Anderegg, P., Bios, 1984,15 (10), 54.

Journal of The Institute of Brewing

This document is provided compliments of the Institute of Brewing and Distilling www.ibd.org.uk Copyright - Journal of the Institute of Brewing